https://bugzilla.kernel.org/show_bug.cgi?id=217400 Bug ID: 217400 Summary: Bugged RMRR prevents PCIe Passthrough of KVM VMs on certain platforms Product: Virtualization Version: unspecified Hardware: Intel OS: Linux Status: NEW Severity: high Priority: P3 Component: kvm Assignee: virtualization_kvm@xxxxxxxxxxxxxxxxxxxx Reporter: bugtracker@xxxxxxxxxxxxx Regression: No Certain vendors that are under the assumption that standards are for jerks and Intel's specifications are a loose optional guideline have implemented RMRR in such a way that every PCI device is marked as reserved and therefore cannot be passed through to a virtual machine. This issue has been very well documented by some people that have a lot more experience than I do at the below linked resource. I was hoping that the upstream Linux kernel could implement the Relaxed RMRR option, enabled by a kernel parameter at boot time, on these bugged platforms as that would re-enable or rather enable a lot of broken servers for the first time ever to use PCIe Passthrough. I can verify the issue exists on a HPE DL360e Gen8 with trying to passthrough a GPU to a KVM/QEMU machine. This bug is very much hardware specific, but seems to only exist on Linux systems as they try to honor the Intel specifications as much as possible, it was reported that other systems (e.g. VMWare ESXi) do not have this issue as they ignore RMRR out-of-the-box. https://github.com/Aterfax/relax-intel-rmrr I can confirm the patched Proxmox-Kernel works fine on an HPE DL360e Gen8. The same bug was also reported to the openSUSE maintainers, they suggested to report it here instead. https://bugzilla.opensuse.org/show_bug.cgi?id=1211069 -- You may reply to this email to add a comment. You are receiving this mail because: You are watching the assignee of the bug.
